Future of Trading & Consumer Apps: ZK & Perps | Thibault Binier - Hibachi, Varun Vruddhula -Hashflow

good afternoon everybody uh my name is tibo B co-founder of Hibachi I'm here with verun uh from from hash flow uh very excited uh to speak with everybody about the future of trading and consumer apps we're going to be talking about ZK we're going to be talking about perpetuals we're going to be uh diving into all those things but uh at the end of the day we're very much focused on the end user and so um yeah uh I think as we as we get started uh one of the things that uh we'd like to lead with is some history it's important to kind of um understand why we're here and how we got here so verun um floor is yours absolutely thank you so much and uh you know uh I'm Veron from haslo so uh uh like Tio mentioned you know to one of the things we want to talk about today is essentially electrical provable Computing essentially uh what we're seeing is a fundamental shift in the way we compute right so where we moving our away from from trusted systems to trustless trustless systems right and as TBO mentioned uh before we talk about this diving a bit into the history of computing is kind of a good way to paint a picture of what's to come right if you think about like history of Technology it's always been about removing trust from the equation like if you think about uh transistors which replaced uh unreliable mechanical switches and giving us comp compute as we know it today then you had the late n early '90s you had tcpip which basically uh removed the need for centralized Gatekeepers making the internet Unstoppable then late '90s we had cloud computing which basically abstracted infrastructure and uh allowed computation to scale and be more accessible to all of us right and then in the 2000s we had the iPhone which basically put all of this in our pockets making compute an extension of ourselves but if you look at all these breakthroughs uh you'll see that they have always always been about removing friction by replacing human discretion with deterministic systems EX systems that uh basically do not systems that do not rely on you know do do not rely on trust why they simply work so keeping this pattern in mind what we seeing now is like you know provable Computing which is like the next era in the evolution the idea is you know how can we run applications that also prove what they're doing without the need to change the entire stack right and years ago uh for years ZK has been considered this uh research breakthrough that wasn't quite ready for production but uh the year's 2025 and it seems like it's finally the year for ZK where the stack is fully ready it's like fast enough it's efficient enough it's mature enough to be the backbone for real world applications right and we're not talking about some theoretical experiments we are talking about fully production ready software that is already transforming the way we operate exchanges protocols you know and various other applications and in this particular instance you know we wanted to discuss how this you know hiachi is one of the exchanges uh that we designed that but basically runs on as a ZK coprocessor the idea being you run a centralized exchange but then you prove itself in seon chain so that at any given time you have mathematical guarantees that the exchange is solvent and you avoid another s SPF like Fiasco right so so uh but to dive more into exchanges you know obviously we're going to talk a bit more about uh hash xos and the infra that we built to enable such exchanges but also would like to you know I mean you've been building exchanges so so uh why don't you tell us more about uh uh history of exchanges and why hiachi is ready for prime time and why is this the moment and why this is needed yeah absolutely I mean it's it's really kind of interesting to see the come up on the on on the compute side of things with ZK and it's also um I think helpful to look at it uh from an application point of view I mean it's no secret to us that trading is probably you know the the core of our industry it's uh very clearly what has the most product Market fit um in in blockchain and crypto so exchanges are really at the foundation of of um of that of everything that we do so just taking a step back and understanding kind of how we got here um we look at the history of exchanges and we can even start actually from traditional finance and and the way that those are uh built and designed and the the bottom line there without getting into too much detail is that there's endless infrastructure endless middlemen and and just it's a very actually opaque uh inefficient system uh so one of the things that I thought you know that I think is very exciting about our industry is that we can build very fast um and trustless exchanges right at least that was the goal um and and the I think it's taken us some time but I think again we'll get we'll get to the point where uh why now is actually uh the real moment in time um so just you know uh the first generation of exchanges in in in crypto were the centralized exchanges and they on boarded millions of users to this day they they carry the most users um out of anything that that that we that we have so that centralized exchange obviously the traits there are going to be um very performant the the ux is usually very good you can kind of trade um seamlessly and and you don't really have to worry about anything um so sure there there's there's that kind of good user experience component but they're very opaque they're blackbox exchanges and the truth is we went from trusting one institution to trusting another one and um you know we don't we shouldn't have to trust uh anybody with our assets I think that's at the ethos of what we're doing so you know uh things like FTX happen and then the decentralized exchange comes about and suddenly everything thing is decentralized and we don't really know why uh other than you know it's it's so that you have control of your assets which that in end of itself is is is a right call to action it's a good value prop but we lost a lot of that performance we lost a lot of that ability to really trade at speed and and with deep liquidity Etc so unfortunately we we um we see a history of uh repeated failures in in terms of our exchanges um until I I think you know uh tying a back to ZK we find that today is actually the right time to build the third generation of exchanges what what we like to call the provable exchange where it's combining everything that you like about the centralized exchange and kind of taking a step back as well and looking back um with with the concept and the notion that the user doesn't care what's decentralized and what isn't at the end of the day they want a good trading experience and they want to make sure their assets are safe and they have that proof of solvency right so that is the third generation of exchanges for us that is the endgame for uh for and and you know also a really good use case for for ZK so um verun I'm going to I'm going to hand it back to you uh to get a little bit uh more context around how we use ZK um in in as an exchange but also just in general how it comes into play to allow for a very good trading experience a fast exchange uh that has that proof of solvency where you don't have to trust another um you know it's it's fully trustless yeah no absolutely I mean one more thing I want to point out here like you when you talked about user experience and Des calization right so but people often over index on security but then I think Hass mentioned a good point the other day uh but the thing we should be asking is security from what right so from whom because you can argue that you know self custody is the most important thing but then not necessarily because when you say self- custody you're basically assuming you're making the trust assumption that the person who custodies the keys is the most secure person to trust you know like they trust themselves with their own security practices which for the most part is not a very good assumption right because most people don't really have very good security practices so in that case if you're telling me that you know you can't really tust yourself with your own Keys which is the case with many many people which is why they prefer institutions some somebody else to do the key management because they're essentially Outsourcing this idea of having secure practices to take care of them right so they're in a way deferring the idea of key security and management to somebody else and in most cases if you're using applications like exchanges it falls on it's it's on the exchanges to have really good practices to do something like this right so and obviously uh the history is a bit murky when it comes to exchange practices and Key Management right so obviously we have had Mount go then we had like um FTX was a big Fiasco byit not necessarily as a bad active but unfortunate but anyway we've seen this over and over right so and this is where I'm like what are we doing as an industry like how do we address this without having to compromise user experience and performance which is essentially Paramount right you can't really build an app and expect a user to learn all these technical inity details which which is almost Irrelevant for most people to use something so that's sort of where the philosophy the design philosophy here is to enable exchanges or even more broadly speaking enable products that do not compromise user experience or performance they operate and and make tradeoffs that directly correlate to a better user experience why you should not be compromising that uh at the cost of anything but once you have that then how do you ensure that you don't get rugged right and this is sort of where you know 5 years ago like I said you could have said like a CK still it's it's infancy it's not quite ready yet we have to do other you know things but it's finally here right so we can finally build systems that where you don't have to trust a centralized human operator to do these things and the thing is the magic of zare here is that it is not fun facing at all so as a user you don't ever need to think about what's going on you simply use the application and and and just do your trades right on the other hand security almost is built in to the application uh by virtue of using ZK right so and that's what we wanted enable with xos and to get a little more technical here I think the way we design the system is if you're an exchange operator you basically run a centralized system you're in a central limit order book but then then what you do is you batch your transactions and you compute a zero knowledge proof on these that basically proves that the state change that happen over time through transactions is always preserved right it's done correctly and then the Z proof itself is published on chain so anybody could verify that the straight transition that happened the way user balances got updated it was done without any any any uh funny business involved uh behind the scenes right so that essentially guarantees uh the Integrity of the order book uh mathematically without without the need for I know like mechy going there and giving an audit report so so anyway ZK is like the mathematical version of meckenzie where where it basically removes the need for audit firms and it basically allows anyone anywhere to independently verify the Integrity of any system not just Financial but over time just the entire digital economy right so that's kind of where we are and obviously uh that's sort of where you know hiachi is one of the first things that's out there where you can really feel the magic of ZK in action without even really seeing ZK which uh it's an interesting matter because ZK is always about you know allowing you to verify things without disclosing things and in an interesting way it kind of works exactly that way but at a more macros scapic level so you know if you want to share anything there yeah absolutely I mean I think it's absolutely right um just the end user doesn't actually need to know that there's any ZK in the background uh we just kind of have to be very clear about the value prop and and what that means for them right so again for us it really just comes down to that proof of solvency and that's actually the first step I think there is a lot of work uh that's still going on to really expand from there on the Privacy side and you know we can we can talk a little bit about that but just it one thing I want to make very clear is that that doesn't neglect the fact that you still have to build a good exchange right so we can talk ZK all day we can say that we do ZK exchange but absolutely I mean without a doubt a good exchange it doesn't matter what kind of complicated stack you have no one is going to use it it's not it's not enough to just say I'm a I'm a ZK exchange and we're going to get so so what makes a good exchange exactly so what what does make a good exchange so at at the core I think for us obviously we we look at success stories and and uh why people trade where they do um and it comes down to a few things right liquidity being a very important one especially if you're trading at size um and then the markets that you have one thing that I would say as well is perpetuals um they're a very exciting asset class that is also native to our industry um I think you know when you look at trafi it's not a thing right they trade options sure but perpetuals are actually very exciting in a way because you can trade with leverage um and just take a directional uh position on on any asset without having to actually hold the underlying asset so um the markets that you list obviously that's number one you have to be able to list very fast you have to be able to have deep liquidity for those markets and then everything else kind of follows right the speed the proof of solvency I actually think those are secondary the the main things that people look for in exchange is like can I trade uh mcoin X on this exchange yes okay this is now being considered as a as a potential viable option there are many exchanges so we all compete on markets we all compete on liquidity so I think there's um you know jumping ahead a little bit in terms of what we're thinking about um at Habachi from from that point of view we want to introduce different types of markets right um the Cycles are getting only faster from inception to death of uh assets which um yeah like like at least last time I was I was making this joke it was a little more fun because at least you had six months before you got rugged now you have to do now rugged an hour you get the full I got desensitized so it's you can imagine as an exchange it's it's a it can be a little bit complicated to operationally kind of be at speed with those things and you look at kind of vaults as Solutions and and kind of decentralized or or or public um liquidity Management Solutions that are out there uh but at the end of the day you have to be able to to list fast uh in order to be a good exchange so um kind of took a little bit of a a side note there but that that to me is what a good exchange means what do you think is it fair to say like essentially liquidity is the king obviously and then speed because many many real serious Traders care about that that sub millisecond latency to go in and out and then obviously like being able to list new markets quickly uh I think those are like the things that an exchange you know cares about and you know and the users of the exchange care about is that accurate yeah exactly I think that's at the top of the list from all the traders that we you know talk to and and there's a lot of kind of behind uh behind this I I do want to before uh we wrap get get a bit on the Privacy side and what what hash flow I was going to ask you actually uh because one of the comparisons I see on Twitter now is like hiachi is a ZK hyper liquid yeah and uh my question is what how do you even defer from hyper liquid what is it that you're offering or V are offering I guess that that that makes it more interesting to even pay attention why should I care the the the key to your question is is actually a lot of it is in the road map um you know we're getting out of a beta right now so the foundation is there um and you know you can go and trade and and uh we we have a running exchange um but from here onwards is when I think the party really starts uh so without over announcing or you know kind of getting in trouble with the with the team um we're looking at things like multic collateral we're looking at lending markets we're looking at mobile very actively um and all those different Market types that um I think are exciting so you know just going back to ZK hyper liquid I think ZK again it's it's the foundation of who we are it's uh really what allows us to give that trustless environment uh and that proof of solvency but from from now on we have to build the hyper liquid you know so the ZK part of it is built and then now we're we're we're getting to the point where we just can focus on what the actual user is going to be you know experiencing yeah I like to think call it pfp basically it's like privacy F you know fast and and then proof of sany basically of pfps so so exactly so basically if things this is in general my criticism around onchain order books to begin with like anytime you say put everything onchain we care about transparency turns out like not everyone wants every all all the transaction data to be on a public Ledger right so because like I don't want anyone to know that like I pay you $10 toight for something I don't know right so and this applies even more relevant you know it's more even more relevant when it comes to like trading when you're trading billions of dollars I don't want my balances to be you know public to anybody right so how do you avoid that like how do you ensure that you know you don't disclose your information and that's one of the reasons why many of the big exchanges why they do proof of reserves where you know the exchange has you know x amount of money but you don't really know if that money is sufficient to meet its liabilities right and one of the reasons why they haven't done it yet is because they can't do it without disclosing user information and uh that is not really good and that's one of the things that ZK actually sorts for is it allows you to prove your s Andy without disclosing user data which is actually a really critical component which which you don't really get unless you're running like you know like an if you don't really get that if you're running a uh onchain order book right so so privacy is actually a critical component there then obviously speed we talked about like you know uh if you're retail maybe one second is not a big deal but if you are somebody who's trading serious money then then you do need that low latency like in the order of milliseconds and then obviously uh knowing that exchange is dolant is not going to pull NS SPF on me that's assuring to say the least yeah absolutely just to wrap up because uh we're coming com more time I I I the last thing I would say is um this is amazing timing I think we're again we're just getting started people are starting to talk about they're starting to look for the next generation of exchanges and uh we've been working at this for a really long time so just generally speaking stay tuned um you know there's lots coming and um yeah excited for it yeah I know so the future just isn't decentralized it is provable future isable exactly all right thank you